Symposium Overview-- June 2nd, 2014 (08:15-18:00)

The Network Science for National Defense Symposium is part of the "NetSci 2014" conference taking place 02-06 June 2014 in Bekeley, CA. The event will be held at the Claremont Hotel and University of Berkeley. For full details visit their site using the logo link on the left.

If interested in attending, please send an email to NetSciNationalDefense@gmail.com

Background

Network Science and other emerging data science methods have proven to be very effective for improving understanding and analysis in U.S. national security and intelligence missions. Because of the nature of these missions however, the benefit from these new analytic methods has not spread in a programmatic way to other mission areas in the Department of Defense. Individual organizations are beginning to use these methods in the social sciences to gain deeper understanding of human behavior for missions like personnel readiness and force management. This symposium will provide an opportunity to exchange these emerging ideas, expose network science practitioners to the important personnel issues in national defense, and allow DoD mission owners to explore the practical application of network science and other methods in addressing the issues facing the U.S. Department of Defense.

Expected outcomes

Each working group is expected to work on a particular pre-defined topic, and articulate potential areas and particular topics for possible research within the DoD, as well as to create collaborative teams to present at next year's symposium.

 Schedule -- June 1st, 2015 (08:30-11:30)

Network Science and other emerging data science methods have proven to be very effective for improving understanding and analysis in U.S. national security and intelligence missions. Because of the nature of these missions however, the benefit from these new analytic methods has not spread in a programmatic way to other mission areas in the Department of Defense. Individual organizations are beginning to use these methods in the social sciences to gain deeper understanding of human behavior for missions like personnel readiness and force management. This symposium will provide an opportunity to exchange these emerging ideas, expose network science practitioners to the important personnel issues in national defense, and allow DoD mission owners to explore the practical application of network science and other methods in addressing the issues facing the U.S. Department of Defense.

 

There will be 3 talks, followed by topic-oriented working groups. Use menu to the left to view Speakers and Topics.

Registration

  • 08:30-08:45 registration and coffee
  • 08:45-09:00 welcome, introductions and overview of the day (Ralucca) and motivation for the meeting (Jon)

Guest Speakers (each talk slot includes 5 min questions/break)

  • 10:00-10:30 Ralucca Gera, Applied Mathematics Department, Naval Postgraduate School (NPS)

              10:30-10:45 Break (snacks & coffee)

  • 10:45-11:15 David Alderson, Director, Center for Infrastructure Defense, Naval Postgraduate School (NPS)
  • 11:15-11:45 Pete Schirmer, RAND
  • 11:45-12:15 Douglas Yeung, RAND
  • 12:15-12:45 Ray Zimmerman, PERSEREC

Break for lunch

Guest Speakers (cont.)

  • 14:00-14:30 Paul B. Lester, Major U.S. Army Research Facilitation Team (RFT)
  • 14:30-15:00 Santiago Gil, Center for Complex Network Research,  Northeastern University,  Boston, MA and Seed Scientific, New York City,  NY 

Topic-oriented Working Groups

  • 15:15-15:30 overview of the topics and expectations of the working group (Mark Breckenridge)
  • 15:30-16:30 group discussions.  Please email Ralucca Gera (rgera@nps.edu) with suggestions for topics to be considered for the working groups

              16:00-16:30 Snacks & coffee

  • 16:30-17:00 preparation of working group deliverables
  • 17:00-18:00 group reports on the findings and results.  Each working group is expected to work on a particular pre-defined topic, and articulate potential areas and particular topics for possible research within the DoD, as well as to create collaborative teams to present at next year's symposium:
    • Summary of the most interesting (controversial) discussion points
    • Potential areas for research in DoD, and specific research topics, if discussed
    • Identification of collaboration teams and topics to be presented at next year's symposium

NetSci open reception (light hor d'oevres, wine and beer)

  • 18:00-20:00 please join us in the Ginko Courtyard of the Clark Kerr campus

 Panel Discussions -- June 3rd, 2014 (08:45-10:00)

The panel discussion will be commencing at 08:45

Panel

The goals of the panels include familiarizing network science practitioners attending the symposium with DoD mission areas that could benefit from network science analytic techniques, especially personnel and force management mission areas that are underrepresented in getting benefit from these tools (e.g. healthcare and other benefit delivery, force accession and attrition analysis) .  Audience participation is encouraged, if you have a question in response to a panel members statement or a follow up question.  The panel might get the attendees creative ideas going with a couple of examples/recommendations of their own for specific areas/topics or methods/techniques that bear further study.

Panelists

Prof. David Alderson
Director, Center for Infrastructure Defense
Naval Postgraduate School

Bio: David Alderson is an Associate Professor in the Operations Research Department and serves as Director for the Center for Infrastructure Defense at the Naval Postgraduate School (NPS). He is also a member of the NPS Cyber Academic Group, which has academic oversight of interdisciplinary cyber curricula on campus.

 Dr. Alderson’s research focuses on the function and operation of critical infrastructures, with particular emphasis on how to invest limited resources to ensure efficient and resilient performance in the face of accidents, failures, natural disasters, or deliberate attacks.  His research explores tradeoffs between efficiency, complexity, and fragility in a wide variety of public and private cyber-­‐physical systems.

 Dr. Alderson has been the Principal Investigator of sponsored research projects for the Navy, Army, Air Force, Marine Corps, and Coast Guard.

Dr. Alderson received his doctorate from Stanford University and his undergraduate degree from Princeton University. He has held research positions at the California Institute of Technology (Caltech), the University of California Los Angeles, the Xerox Palo Alto Research Center (PARC), and the Santa Fe Institute. He has extensive industry experience and has worked for several venture-­‐back startup companies. His early career was spent developing technology at Goldman Sachs & Co. in New York City.

 

Prof. Chris Arney
Chair of Network Science
United States Military Academy

Bio: Chris Arney is a Professor of Mathematics at the United States Military Academy.   His PhD is in mathematics from Rensselaer Polytechnic Institute.  Chris spent most of his 30-year military career as a mathematics professor at USMA.  He also served as Vice President for Academic Affairs at the College of Saint Rose in Albany and division chief of the Mathematical Sciences Division of the Army Research Office (NC), where he managed and performed research in cooperative systems, information networks, pursuit-evasion modeling, intelligence processing, artificial intelligence, and language for robots.  Chris has authored 22 books, written 150 technical articles, made over 250 presentations, and reviewed over 200 books. His technical areas of interest include mathematical modeling, network science, and cooperative systems

 

Dr. William G Hanley
Senior Managing Scientist
Exponent

Bio:  Dr. William Hanley is a Senior Managing Scientist within Exponent’s Technology Development practice. With over thirty years of experience in applied mathematics, statistics and computer science, Dr. Hanley has expertise in machine learning, data mining, decision methodologies, systems modeling/analysis and uncertainty quantification. Dr. Hanley has served as the principal investigator and/or co-investigator on numerous applied research and advanced development efforts focused upon: threat/anomaly detection, text mining, cyber security, infrastructure protection, situation assessment, complex system modeling, and environmental remediation. These efforts have resulted in 8 Patents/ROIs and 50+ refereed publications.

Prior to joining Exponent, Dr. Hanley held multiple leadership positions at Lawrence Livermore National Laboratory (LLNL) including Deputy Division Leader for the National Security Engineering Division, Section Leader for Systems and Intelligence Analysis and Focus Area Leader for Engineering Systems for Knowledge and Inference. Before joining LLNL, Dr. Hanley managed the Scene Analysis Department and served as a Senior Scientist at Advanced Decision Systems. Earlier, while at Hughes Aircraft Company, he led the Algorithm Development & Analysis Section and the Mathematical Analysis Group.

Dr. Hanley holds PhD in Applied Statistics from the University of California, Riverside and an M.S. degree in Computer Science from the University of California, Los Angeles. He is an active member of the American Statistical Association

 

Dr. Eric Lang
Director, Personnel & Security Research Center (PERSEREC)
Defense Manpower Data Center (DMDC), U.S. Department of Defense

Bio:  Dr. Eric Lang is the Director of the Defense Personnel and Security Research Center, PERSEREC – a Division of DoD’s Defense Manpower Data Center and has 30 years of experience designing and managing social science research to improve national policies and programs. He received his doctorate (Ph.D.) in social psychology from the University of Michigan where he focused on experimental and survey research at the Institute for Social Research, followed by several years at the American Institutes for Research working on national projects related to education, research methods and health. Prior to joining PERSEREC, Dr. Lang was a Principal Research Scientist and Division Director at Sociometrics Corporation where he managed research on federal grants and contracts related to the development and use of national data archives, social science applications for the Internet, and research methodology.

Dr. Lang has worked at PERSEREC since 2000, designing and directing basic and applied research to improve the effectiveness, efficiency and fairness of personnel security and suitability systems (e.g., security clearance programs) and policies for the Department of Defense (DoD) and nationwide. Outside of DoD, Dr. Lang has provided research consultation to the Department of State, Department of Homeland Security, Department of Energy, Department of Health and Human Services and the Intelligence Community. As a member of the federal Joint Security and Suitability Reform Team, he helped to improve personnel security and employment suitability operations, information systems and policies nationwide.

Examples of implemented initiatives for which Dr. Lang played a key role include:

  • Clean-case automated electronic security clearance adjudication (“eAdjudication”).
  • A method for assessing security clearance adjudication quality: “Review of Adjudication Documentation Accuracy and Rationales” (RADAR).
  • The “Counterintelligence Reporting Essentials” (CORE).
  • The “Dispositional Indicators of Risk Exposure” (DIRE) scale for assessing personality disorders that are security/reliability risks.

Additional areas of research interest include

  • Data science and Big Data approaches to addressing personnel problems and opportunities.
  • Assessing “CyberPsychology” risks to workplace reliability and performance.
  • Assessment and risk management of insider threat (e.g., espionage and workplace violence).

Dr. Lang has been an invited speaker at national and international social science meetings and government conferences; authored numerous technical publications, conference papers, journal articles, and book chapters; and holds membership in the Association for Psychological Science, Society for Personality and Social Psychology and the American Psychological Association.

 

Panel Facilitator

LTC Jonathan Alt
U.S. Army
Naval Postgraduate School

Leading questions:

  • What is network science and why should the DoD care about it?
    • What are its strengths and weaknesses?
    • What limits its application?
    • Characterize a problem that is the ideal case for the application of network science?
  • Can you describe a successful application of network science to a DoD problem?
    • What were the challenges associated with this application?
    • Are these challenges unique to DoD?
  • Can you describe an example of an inappropriate application of network science to a DoD problem?
    • What drove the application of network science to this particular problem?
  • How does network science relate to data science and "Big Data"?
    • Does network science require similar skill sets?
    • What unique skills are required of network scientists?
  • How would you apply network science to the challenge of improving the ability of Navy recruiters to reach Millennials to generate prospects for service in the Navy?
    • What data would you need?
    • What challenges would come up?
    • Would this be a good application area?
  • How would you apply network science to reduce attrition of Army personnel following their initial term of service?
    • What data would you need?
    • What challenges would come up?
    • Would this be a good application area?
  • How would you apply network science to facilitate management of healthcare providers and facilities?
    • What data would you need?
    • What challenges would come up?
    • Would this be a good application area?

 

 Guest Speakers and Topics -- June 3rd, 2014 (10:00-15:00)

A Large-Graph Comparison Measure for the topology of the internet 

Speaker: Prof. Ralucca Gera, Applied Mathematics Department, Naval Postgraduate School (NPS)
Time: 10:00-10:30

Abstract: This presentation will build on The Seven Bridges of Königsberg, the problem that is at the origin of graph theory, posed by Leonhard Euler in 1735. The talk will present a natural extension of this problem to today's type of problems that still need solutions: mapping complex networks. We will explore one particular complex network, the Internet, and share our recently obtained results comparing snapshots of Internet’s topology. For our research, we modeled the Internet by a graph, with router interfaces represented as vertices, and the connections between these interfaces -obtained through traceroutes- represented by the edges. The modeling of the internet by a graph will be described, but the focus of the talk is on the study of the graphs.

 

10:30-10:45 Break (snacks & coffee)

Network Science for Military Applications:  Practice, Pitfalls, and Potential

Speaker: Prof. David Alderson, Director, Center for Infrastructure Defense, Naval Postgraduate School (NPS)
 Time: 10:45-11:15

Abstract:  Recent efforts to develop a universal view of complex networks have created both excitement and confusion about the way in which knowledge of network structure can be used to understand, control, or design system behavior.  In this talk, I will summarize some of the most prevalent pitfalls in the application of network science to practical problems, and I will comment on the opportunities for contribution in Department of Defense applications.  I will explicitly connect this perspective to the recent explosion in the analysis of social media and other large, unstructured data.

Using Intelligent Applications to Improve Talent Management

Speaker: Pete Schirmer, RAND Time: 11:15-11:45

Abstract:  Intelligent applications that incorporate machine learning and data mining methods can improve the ways in which large organizations manage their people.  Such an approach has several benefits.  It is passive and unobtrusive, as opposed to a survey- or expert-based approach that requires intensive human participation.  Many processes can be automated and maintained inexpensively, so that data collection and analysis can continue well beyond the life of the project.  The analysis need not be constrained by a pre-established set of factors (e.g., a fixed set of articulated competencies).  Lastly, such an approach presents a “ground truth” picture of skills, experience, education, and training that are in demand, where that demand exists, and in what magnitude.  RAND created a Talent Management Application to cluster senior DoD civilian personnel using data and text mining algorithms.  In phase 1 of the project, we clustered Senior Executive Service personnel based on a corpus consisting of resumes, curricula vitae, and job position descriptions.  The analytic engine is written primarily in Java and relies heavily on Free and Open Source Software from the Apache Software Foundation.  Its three main components for data processing are Mahout, Hadoop, and Lucene.  Sitting on top of the analytic engine is a browser-based interface to enable users to explore the clusters graphically and provide search capability.  This paper will provide a high-level description of how the key pieces fit together, how we gathered our data and how it might be collected on a continuous and more-sustainable basis, and some possible uses for this application within Department of Defense organizations

Social Media and Women in Military

Speaker: Dr. Douglas Yeung, RAND Time: 11:45-12:15

Abstract:  The recent announcement of the opening of all occupations to women and the publicity surrounding sexual assaults in the military are media messages that may influence public perception of military service. Women themselves are likely to be affected directly by media messages and influencers of women, including their friends, counselors, and family members are likely to be affected by these media messages. Together, these effects will be felt in changes in recruiting women for military service, such as affecting recruiters in unknown ways. Social media provides an opportunity to explore naturally occurring discussions and relatively unfiltered attitudes regarding women in the military. Social media content may therefore serve as a useful complement to existing measures of attitudes toward the military and propensity to serve.  This research explores how analysis of social media can provide a leading-edge perspective on how the public feels regarding women serving in the military. The purpose of this work to understand the content and tone of media messages that are transmitted via social media where military service is being discussed, beginning with Twitter. We use data mining techniques to collect and analyze relevant tweets. We then report on analyses of social media messages regarding the service of women in the military, describing themes that may be suggestive of recruiting trends or propensity to serve.

Building a New Data Science Capability for the Defense Manpower Data Center

Speaker: Dr. Ray Zimmerman, Personnel & Security Research  Center (PERSEREC) Time: 12:15-12:45

Abstract:  The Defense Manpower Data Center was initially established to provide statistical answers to manpower and personnel questions for the Department of Defense.  While the agency's mission has grown to support additional needs of the Department, analytic approaches have been relatively static. A new initiative is underway to combine technological advances in handling massive amounts of data from sources, standard statistical methods, and machine learning approaches. The goal is to answer questions that cannot be answered today with current approaches. This session will describe initial efforts in building a Hadoop cluster, proof of concept demonstration projects, and future directions.

Lunch break

The Person-Event Data Environment: Leveraging big data for studies of psychological strengths in soldiers.

Speaker: Maj. Paul B. Lester,  U.S. Army Research Facilitation Team (RFT) Time: 14:00-14:30

Abstract:  The Person-Event Data Environment (PDE) is a safe and secure data analysis environment where data from disparate systems from across the Department of Defense (DoD) is integrated and analyzed for research purposes.   The PDE not only houses over 250 personnel data assets from across DoD but also a) hosts a wide range of statistical analysis tools for use by researchers; b) incorporates a robust human subjects protections governance structure; and c) gives researchers access to the Research Facilitation Team, an organization tasked with both easing the burden of accessing the data and providing them with meta data about the PDE’s holdings.  We will begin this discussion with a review of the PDE system and then transition to how the PDE is being used now to provide DoD leaders with a greater understanding of the psychological health and resilience of service members.  Finally, we will outline how the PDE is being postured to “go to scale” over the next year and chart a few future directions for research.     

A Genetic-epidemiology Approach to Cyber-security.

Speaker: Santiago Gil, Center for Complex Network Research,  Northeastern University,  Boston, MA and Seed Scientific, New York City,  NY     Time: 14:30-15:00

Abstract:  While much attention has been paid to the vulnerability of computer networks to node and link failure, there is limited systematic understanding of the factors that determine the likelihood that a node (computer) is compromised. We therefore collect threat log data in a university network to study the patterns of threat activity for individual hosts. We relate this information to the properties of each host as observed through network-wide scans, establishing associations between the network services a host is running and the kinds of threats to which it is susceptible. We propose a methodology to associate services to threats inspired by the tools used in genetics to identify statistical associations between mutations and diseases. The proposed approach allows us to determine probabilities of infection directly from observation, offering an automated high-throughput strategy to develop comprehensive metrics for cyber-security.

Following our symposium, NetSci will be sponsoring an open reception between 6:00-8:00 PM.  We hope you will all join us at this reception to continue the good discussions from the symposium as well as solidify the collaborative efforts that emerge.

 Working Groups

Each working group is expected to work on a particular pre-defined topic, and articulate potential areas and particular topics for possible research within the DoD, as well as to create collaborative teams to present at next year's symposium. 
A technical report summarizing the discussions of the working groups can be found here.

 

 Program Co-Chairs

 

 Technical Report

A technical report summarizing the discussions and presentations can be found here
A publication relating this symposium to other DoD's Network Science symposia can be found
here.

 Funding Opportunities

The DARPA Information Innovation Office (I2O)  runs a number of programs that generate grants.  They are all listed here: http://www.darpa.mil/Our_Work/I2O/Programs/  

Some of the specific efforts related to "big data" are: 
   
XDATA
http://www.darpa.mil/Our_Work/I2O/Programs/XDATA.aspx
    Big Mechanism http://www.darpa.mil/Our_Work/I2O/Programs/Big_Mechanism.aspx
    Anomaly Detection at Multiple Scales (ADAMS) http://www.darpa.mil/Our_Work/I2O/Programs/Anomaly_Detection_at_Multiple_Scales_(ADAMS).aspx
    Detection and Computational Analysis of Psychological Signals (DCAPS) http://www.darpa.mil/Our_Work/I2O/Programs/Detection_and_Computational_Analysis_of_Psychological_Signals_(DCAPS).aspx
    Durkheim Project
http://www.durkheimproject.org/  
    Office of Naval Research (ONR)http://www.onr.navy.mil/
    Air Force Office of Scientific Research (AFOSR)
http://www.wpafb.af.mil/afrl/afosr/  
    Army Research Lab (ARL)http://www.arl.army.mil/www/default.cfm

 

 Symposium's Proceedings

A technical report summarizing the discussions and presentations can be found right here.  A publication relating this symposium to other DoD's Network Science symposia can be found here.

 


[Top]